About cohorts
Examine the behavior and performance of groups of users related by common attributes.
A cohort is a group of users who share a common characteristic that is identified in this report by a Google Analytics dimension. For example, all users with the same Acquisition Date belong to the same cohort. The Cohort Analysis report lets you isolate and analyze cohort behavior.
In this article:
See cohort data
Ways to use cohort data
Next steps
See cohort data
The Cohort Analysis report is available for properties using Universal Analytics. No changes to the tracking code are necessary.
To see cohort data:
Sign in to Google Analytics.
Navigate to your view.
Select the Reporting tab.
Select Audience > Cohort Analysis.
Ways to use cohort data
Cohort analysis helps you understand the behavior of component groups of users apart from your user population as a whole. Examples of how you can use cohort analysis include:
Examine individual cohorts to gauge response to short-term marketing efforts like single-day email campaigns.
See how the behavior and performance of individual groups of users changes day to day, week to week, and month to month, relative to when you acquired those users.
Organize users into groups based on shared characteristics like Acquisition Date, and then examine the behavior of those groups according to metrics like User Retention or Revenue.
Organise users into groups, and compare group performances.
The Cohort Analysis report lets you get a more accurate picture of your relationship with users. With this report, you can organise users into groups based on shared characteristics, like acquisition date. This lets you analyse and compare the behaviour and performance of different cohorts across a variety of metrics, like User Retention and Revenue.
Understanding cohorts
Cohorts are sets of users organised into groups united by a common element and a time frame. In the Cohort Analysis report, this element is a Google Analytics dimension or an attribute of the users. The time frame can be a Day, Week or Month. For example if the cohorts are organised by the dimension Acquisition Date and the time frame Day, then all users in each cohort started their first session on the same day.
Cohorts let you see how users in one group compare to each other, and it also lets you compare behaviour across different groups. Use this report to help you better understand which app versions, which features and content, and which ad campaigns attract more long-term and more frequent users.
For example, you could use this data to see if the percentage of users that had their first session last month continue to return to your content more (or less) than users that you first acquired this month.
Reading the Cohort Analysis report
You can find the Cohort Analysis report in the Audience section of a reporting view.
You can only analyse one dimension and one metric at a time in this report. Use the Cohort Type menu to select the dimension, and the Metric menu to select a metric.
Rows represent cohorts. The size of the cohort is listed as a whole number in the margin of each row in the Cohort Type column in the data table.
Columns display the Cohort Type and the number of days that have passed since the event that is the organising element of the cohorts. Day 00 is the day on which all members performed an action to be grouped together. For example, if the cohorts are based on the Acquisition Date, Day 00 is the day all users started their first session, Day 01 is the day after the first session happened, etc.
Cells display the number of users that belong to that cohort and the corresponding time period. The colour intensity in each cell visually indicates the percentage of users relative to the total population of that cohort. A higher percentage is represented by a darker shading. You can also hover over a cell for details.
The top row of the table displays the average retention across All Cohorts. Any changes that you make to the data view (like changing the Cohort Date Range) are reflected in that average.
You can change the data displayed in the report by selecting different options from the menus that appear above the data table.
This report supports both default and custom Advanced Segments. Each Segment appears as a separate data table in the report. You can compare up to four segments at a time
Configure the report
Use the menus to select:
- The dimension that characterizes the cohorts (Cohort Type)
- The size of the cohorts (Cohort Size): You determine the size of the cohort by selecting the value type for the dimension. For example, if you determine the cohort by the dimension Acquisition Date, you can change the dimension value type to day, week, or month. With these settings, a cohort would be all users who were acquired on the same day, or during the same week or month.
- The metric you want to evaluate (Metric)
- The relative date range of the data, and the number of cohorts (Date Range)
- Which cohorts are illustrated in the chart (N selected)
Understand the data
Chart
By default, the chart shows the cumulative metric values for all cohorts. Use the N selected menu to select a cumulative chart line and/or chart lines for individual cohorts.
Columns
The first column identifies the cohorts and the number of users in each cohort. For example, if the dimension by which you characterize the cohorts is Acquisition Date, this column lists the acquisition date for each cohort, and the number of users you acquired during that time frame (day, week, month).
The rest of the columns reflect the time increments you choose for Cohort Size. For example, if you select by day, then each column includes one day of data. There are 13 time-increment columns, 0-12.
Rows
The first row shows the total metric value for all cohorts for each column. For example, if the metric is Pageviews and the columns are daily data, then the first row shows the total pageviews for the day.
The other rows show the values for the individual cohorts.
Cells
The cells for time increments 0-12 hold the relevant metric values. For example, if you are using the Pageviews metric, then each cell contains the number of pageviews per cohort per time increment.
Segments
When you apply Segments to this report, the data for each Segment is displayed in a separate table.
Examples
Micro trends
Examining the micro trends that in aggregate constitute your macro trends can give you a more realistic picture of your business. For example, your quarterly data might show a steady increase in transactions over that period, which you would regard as a positive outcome. If, however, you were to examine the weekly cohorts that make up that larger data set, you might find that while an overall influx of new users is contributing to a growing number of transactions, there is a regular, dramatic decline in transactions after week 5. Now you know exactly when to reengage users (week 4) in order to improve the performance of each micro trend, and thereby multiply the effect on your macro trend.
Consistency, improvement, or deterioration across cohorts
By simply comparing the values in a single column, you can see whether there’s consistent behavior among your cohorts, or whether performance improves or deteriorates. As you look down the column at data for each newer cohort, you’re looking forward in time (for example, Day 5 for the second cohort occurs after Day 5 for the first cohort though they appear in the same column).
If you’re evaluating daily data, you can look at a single column, say the Day 5 column, to see whether all cohorts perform at about the same level at that point in their experience, or whether the data indicate improving or deteriorating trends. For example, if you are retaining the same percentage of users across all cohorts at Day 5, then that can indicate a comforting consistency in user experience. On the other hand, if you see a steady increase in retention at Day 5, you might be able to correlate that with an improvement in your content or an upgrade to the speed at which your app performs. A steady deterioration of user retention at Day 5 might indicate stale content, or an unusually difficult or poorly coded level in a game--something that is causing fewer and fewer users to continue with the experience.
Engagement, retention, and acquisition
Understanding the point at which users tend to disengage (for example, initiate fewer sessions, view fewer pages, generate less revenue) can help you identify two things:
- Common points of attrition that might be easily remedied
- The rate at which you need to acquire new users to compensate for unavoidable attrition
For example, if you notice that revenue regularly starts to decline in the third or fourth week after acquisition, you might reengage users with a remarketing or email campaign that offers discounts or ads for new products that have been added since their last sessions. You could also reengage those users with dynamic remarketing by offering ads for products related to the ones they purchased during their initial engagement.
If you identify inevitable patterns of attrition, say 10% a month, then you are able to understand the rate at which you need to acquire new users to create the growth rate you want for your business.
Response to short-term marketing efforts
If you run short-term marketing efforts like single-day email campaigns, this report gives you the chance to track the behavior of just the users you acquired during the related time frames. For example, if you’re running successive 30%-off, 25%-off, and 20%-off campaigns as a holiday approaches, you can see how different metrics like Revenue per User andTransactions per User compare among the groups of users you acquired on the dates each campaign ran.
Content From Google.com
0 comments:
Post a Comment